The research team, led by Ciara O’Sullivan and Tara Coppinger along with Indika Dhanapala, Diarmuid Boyle and Alison Merrotsy, has built MindBia around a co-design philosophy. Rather than developing an app in isolation and hoping it fits, the researchers drew on prior studies with both service users and mental health staff to shape the content and delivery. The result is a structured program of six modules delivered through a web-based application, aimed at individuals residing in high-support mental health hostels. That setting matters: hostels are home to some of the people most affected by the mortality gap, yet they are rarely the focus of digital health research, which tends to recruit from outpatient clinics or the general population.

The scientific rationale for targeting nutrition in severe mental illness is well established. Epidemiological studies consistently show that cardiovascular disease, type 2 diabetes and metabolic syndrome occur at dramatically elevated rates in people with diagnoses such as schizophrenia and bipolar disorder, and that dietary patterns are a major modifiable contributor. Traditional face-to-face dietary counselling, however, is resource-intensive and often impractical to deliver at scale within stretched mental health services. Digital interventions offer a potential solution because they can be scaled, accessed flexibly, and updated cheaply. The catch is that very few digital programs have treated nutrition as the primary outcome for people with severe mental illness; most target physical activity, smoking or medication adherence, or fold diet into a broader lifestyle package.

MindBia’s pilot study is explicitly framed as a feasibility and acceptability trial, not a test of effectiveness. This distinction is central to how modern behavioral intervention science works. Before investing in a large randomized controlled trial, researchers must establish whether the intervention can actually be delivered as intended, whether participants will engage with it, whether the assessment procedures work in practice, and whether the target population finds the program usable and worthwhile. Skipping this stage is one of the most common reasons promising health technologies fail in later trials: recruitment stalls, retention collapses, or engagement with the digital tool turns out to be far lower than anticipated.

Methodologically, the pilot employs a mixed-method pre-post design over eight weeks. Participants will complete assessments at baseline and again after the intervention period, covering dietary behaviours, digital literacy and general wellbeing. Measuring digital literacy at the outset is a thoughtful addition, since it acknowledges that any web-based intervention for this population must contend with wide variation in confidence and skill with technology. Dietary assessment in this group is notoriously difficult, and capturing baseline digital capability helps the team interpret engagement data and identify who might need additional support to use the application at all.

Engagement will be monitored continuously through embedded usage analytics within the web application itself. This is one of the quiet advantages of digital interventions over traditional programs: every login, module completion and interaction can be logged automatically, producing an objective record of how the intervention is actually being used rather than relying solely on self-report. For a feasibility study, these analytics are arguably as important as the outcome measures, because they reveal whether participants open the app, how far they progress through the six modules, and where in the program they drop off. Patterns of non-use can point to specific design flaws that would otherwise remain invisible until a full trial had already failed.

The qualitative arm of the study adds another layer of insight. After the eight-week intervention period, participants will take part in semi-structured interviews exploring their experiences of the program, its perceived impact on their eating and wellbeing, and the barriers they encountered along the way. In a population where trust in services and technology can be fragile, and where cognitive and motivational difficulties associated with mental illness can complicate engagement, these first-person accounts are essential. Quantitative analytics can show that someone stopped using the app in week three; only the participant can explain whether that was because of a design problem, a change in their mental state, a practical obstacle in the hostel environment, or something the researchers never anticipated.

The choice of a pre-post design rather than a randomized comparison is appropriate for this stage of development. Pre-post studies measure outcomes before and after the intervention within the same group of participants, providing preliminary signals about potential changes in diet and wellbeing without the statistical power or causal rigor of a controlled trial. The researchers are careful about what such data can and cannot show. Any improvements observed could reflect the intervention, but also natural fluctuation, attention from researchers, or other changes in participants’ lives. That is precisely why the pilot’s stated purpose is to inform refinement of the MindBia application and to guide the design of a larger-scale trial that can properly evaluate effectiveness.

The co-design heritage of MindBia deserves particular attention in the broader context of digital health equity. There is growing concern that digital interventions, left unexamined, can widen health inequalities because they work best for people who are already motivated, literate and connected. People with severe mental illness are disproportionately likely to be on the wrong side of that divide, facing lower rates of employment, lower incomes and higher rates of digital exclusion. By involving service users and hostel staff from the earliest stages, and by explicitly measuring digital literacy as part of the pilot, the research team is attempting to build an intervention that meets people where they are rather than assuming a level of capability and motivation that many do not have.

If the pilot demonstrates that MindBia can be delivered, that people in high-support hostels will engage with it, and that the measurement procedures are workable, the path opens toward a definitive trial of effectiveness, and eventually toward a scalable tool that mental health services could deploy widely. If it reveals problems, the mixed-method design should show exactly where they lie and how to fix them. Either way, the study contributes to a small but growing body of work insisting that the physical health of people with severe mental illness deserves the same methodological rigor and technological creativity that other areas of digital health have received. Closing a mortality gap measured in years will not happen through a single app, but interventions like MindBia represent the careful, unglamorous groundwork, testing feasibility one module at a time, on which any meaningful progress will have to be built.

Subject of Research: Feasibility of a web-based nutrition intervention for people with severe mental illness

Article Title: A pre-post pilot feasibility study of the MindBia web-based nutrition intervention for individuals with severe mental illness: Study protocol

Article References: O’Sullivan, C., Coppinger, T., Dhanapala, I., Boyle, D., & Merrotsy, A. (2026). A pre-post pilot feasibility study of the MindBia web-based nutrition intervention for individuals with severe mental illness: Study protocol. PLOS Digital Health, 5(9), e0001756. https://doi.org/10.1371/journal.pdig.0001756

Image Credits: AI Generated

DOI: 10.1371/journal.pdig.0001756

Keywords: severe mental illness, nutrition intervention, digital health, MindBia, feasibility study, co-design, web-based application, high-support hostels, diet quality, mixed methods, health inequalities, pilot study